1. Field of Invention
The present invention relates to methods for manufacturing membrane electrode assemblies useful in proton exchange membrane and direct methanol fuel cells, including compositions useful for the deposition of the various layers in the membrane electrode assemblies. The methods enable rapid production of the membrane electrode assemblies and layers used therein. The present invention also relates to membrane electrode assemblies, including the gas and fluid distribution layers used therein, as well as proton exchange membrane and direct methanol fuel cells produced by the method.
2. Description of Related Art
Fuel cells are electrochemical devices that are capable of converting the energy of a chemical reaction into electrical energy without combustion and with virtually no pollution. A typical fuel cell includes a stack of membrane electrode assemblies (MEAs). Generally, MEAs comprise an anode, a cathode and a solid or liquid electrolyte disposed between the anode and the cathode. Different types of fuel cells are categorized by the electrolyte used in the fuel cell, the five main types being alkaline (including metal-air fuel cells), molten carbonate, phosphoric acid, solid oxide and proton exchange membrane (PEM) or solid polymer electrolyte fuel cells (PEFCs). A particularly preferred fuel cell for portable applications, due to its compact construction, power density, efficiency and operating temperature, is a proton exchange membrane fuel cell (PEMFC) which can utilize methanol directly without the use of a fuel reformer. This type of fuel cell is referred to as a direct methanol fuel cell (DMFC). Other liquid fuels may also be used in a liquid fueled fuel cell including formic acid, formaldehyde, ethanol and ethylene glycol. DMFCs are related to PEMFCs because both types employ a proton exchange membrane and oxygen as an oxidant on the cathode, but they are different in that DMFCs utilize liquid methanol as the fuel on the anode whereas a PEMFC utilizes a gas feed containing hydrogen. The hydrogen feed for a PEMFC can originate from a reformed methanol fuel feed as well as other fuel sources such as methane or propane. DMFCs are attractive for applications that require relatively low power because the anode achieves the combined function of “reforming” the methanol directly into hydrogen ions that can be delivered to the cathode through the PEM. This avoids the need for a separate methanol reforming system thereby reducing the size, and potentially the cost, of the overall system.
A DMFC is comprised of membrane electrode assemblies (MEAs). A cross-sectional view of a typical MEA is illustrated in FIG. 1 (not to scale). The MEA 100 comprises a PEM 102, an anode 104 and a cathode 110, and anode and cathode fluid distribution layers 106 and 112. The anode and cathode each include an electrocatalyst layer 120 and 122 which sandwich the PEM. The gas or fluid distribution layers, 106 and 112, typically include a carbon-based substrate and each layer is located next to its respective electrode. Bipolar plates 108 and 114 are disposed between the anode and cathode of sequential MEA stacks and comprise current collectors and flow fields, 116 and 118, for directing the flow of incoming reactant fluid to the appropriate electrode. Two end plates (not shown), similar to the bipolar plates, are used to complete the fuel cell stack.
The bipolar plates serve as electrically conductive separator elements between two adjacent MEAs, and (1) have reactant distributing grooves on both external faces thereof, (2) conduct electrical current between the anode of one MEA and the cathode of the adjacent MEA in the stack, and (3) in most cases, have internal passages which are defined by internal heat exchange faces and through which coolant flows to remove heat from the stack. Bipolar plates are often fabricated from graphite which is lightweight, corrosion resistant and electrically conductive in the PEM fuel cell environment. In certain cases, bipolar plates may also be fabricated from metals including expanded metals, metal foams and porous metal sheets. Bipolar plates are also gas impermeable to prevent mixing of the two reactants, which would lead to direct oxidation of the methanol. Bipolar plates are described in more detail in, for example, U.S. Pat. No. 5,776,624 by Neutzler and U.S. Pat. No. 6,255,012 by Wilson, et al, which are incorporated herein by reference in their entirety.
During operation of the DMFC, methanol is supplied to the anode and oxygen (air) is supplied to the cathode to create the reactants necessary to operate the fuel cell. Methanol flows through the flow fields 116 of bipolar plate 108, through the anode fluid distribution layer 106 and to the anode electrocatalyst layer 120, where it is oxidized. Oxygen flows through the flow fields 118 of bipolar plate 114 through the cathode fluid distribution layer 112, and to the cathode electrocatalyst layer 122 where the oxygen molecules are reduced to oxygen ions. Electrons from the oxidized methanol are routed to the cathode 110 through an external circuit 130 connecting the bipolar plates 108 and 114 to produce electrical current. Protons from the oxidized methanol pass from the anode to the cathode through the PEM 102 and recombine with the electrons and ionized oxygen to form water. Carbon dioxide is produced at the anode and is removed through the exhaust of the cell. The foregoing reactions can be written as follows:Anode: CH3OH+H2O→CO2+6H++6e−  (1)Cathode: 6H++6e−+ 3/2O2→3H2O  (2)Overall: 2CH3OH+3O2→2CO2+6H2O+energy  (3)
Although the theory behind fuel cell operation has been known for over 100 years, there has been difficulty producing commercially viable fuel cells due to technological barriers, and also due to the availability of more cost-effective energy sources such as petroleum. However, devices using petroleum products, such as the automobile, produce significant pollution and may eventually become obsolete with the depletion of petroleum resources. As a result, there is a need for an alternative means for producing energy. Fuel cells are a promising alternative source of energy in that they are relatively pollution-free and utilize hydrogen, a seemingly infinite fuel source.
Among the critical issues that must be addressed for the successful commercialization of fuel cells are cell cost, cell performance and operating lifetime. For stationary applications, improved power density is critical. For automotive applications, high voltage efficiencies are necessary. In terms of cell performance and operating lifetime, it is important that the fuel cell be constructed to minimize kinetic, ohmic and mass transport losses within individual MEAs. A major technical challenge is enabling the efficient transport of methanol, water, oxygen, carbon dioxide, protons and electrons to and from the relevant locations in the MEA. The efficient transport of these reactants and by-products will increase both the cell performance and the operating lifetime of the fuel cell. Most current manufacturing methods are not capable of forming structures that enable all around efficient transport of reactants and by-products within the MEA.
Aside from optimizing the cell performance by improving the aforementioned transport characteristics of the MEA, the successful commercialization of fuel cells depends on producing MEAs at a reasonable cost. The ideal MEA manufacturing process is one that rapidly and continuously produces MEAs having good transport characteristics, and with reduced manufacturing down-time and reduced capital costs associated with the MEAs. One way to decrease capital costs, without decreasing cell performance, is to place the required materials only where needed within the MEA structure, especially the expensive catalyst materials.
Fuel cells operate most efficiently when the electrocatalytically active sites are in direct contact with both the PEM and the supply fuel. This location within the electrode is known as the three-phase interface. Most conventional methods for manufacturing MEAs cannot deposit catalyst and other materials in predetermined locations to increase the catalyst utilization. As a result, an unnecessarily large amount of catalyst is deposited. Depositing the catalyst in relation to other materials such that the catalyst is located only in the areas where the three-phase interface (discussed below) occurs would decrease the capital cost of the MEA without sacrificing performance.
There are many different methods for producing the layers of an MEA including physical vapor deposition, screen printing, dry powder lamination, spraying, extrusion (a.k.a., slot die), electrostatic printing, and dry powder vacuum deposition. However, there are many drawbacks to these manufacturing methods. Many of the processes are laboratory methods, which cannot be used in full-scale manufacturing processes. Many of the processes are slow and require significant amounts of down-time to change patterns. Many of the processes cannot directly deposit materials on flexible or low-temperature surfaces such as thin polymer membranes. Many of the processes cannot precisely control the volume or concentration of the deposited materials on selected microscopic areas of the substrate and many of the processes cannot deposit the materials on the substrate in a pattern or cannot deposit patterns with a small line width. As a result, controlled gradients in composition, concentration or porosity within layers are difficult to fabricate. Many of the processes are not compatible with all of the required materials which results in the use of different deposition methods during the fabrication of the MEA. This increases the number of deposition steps, which in itself is inefficient and may lead to alignment errors within the MEA and lead to unnecessary handling and breakage of fragile components such as the thin PEM.
Lamination is well-known in the art and in the context of MEA manufacture refers to a process where two components, e.g., a PEM and an electrode, are bonded together using heat, pressure and/or an adhesive. U.S. Pat. No. 6,197,147 by Bonsil et al. discloses a laminating process for producing MEAs. The process comprises laminating an ion-conductive membrane, a catalytically active substance and a gas permeable electron conductive material. It is disclosed that the membrane and/or conductive material can be contacted with a compound such as hexachloroplatinic acid, which is then reduced to form platinum metal.
Decalcomania is generally defined as a process for transferring designs printed on a specially prepared transfer substrate to materials such as glass or metal. In the context of MEA assembly, the designs are the deposition patterns used to create the various layers of the MEA which are then transferred to a MEA substrate, such as a PEM, gas or fluid distribution layer or bipolar plate. In practice, an ink composition is deposited onto a blank cartridge, herein referred to as a transferring substrate. The transferring substrate and the MEA substrate are then sandwiched together and pressed in a heated environment. The heat cures the ink to the substrate thereby releasing it from the transferring substrate and depositing it on the substrate. The transferring substrate is then removed by peeling it from the deposited ink layer.
U.S. Pat. No. 5,211,984 by Wilson discloses a method of producing MEAs by decalcomania. An ink composition comprising a supported platinum catalyst and a proton conducting polymer is painted on a release blank to form a thin film decal. The decal is then transferred to the surface of a PEM using a heat press.
Decalcomania and lamination are advantageous in that they can deposit materials on fragile surfaces and are relatively simple to operate. Lamination also provides a high throughput. However, decal and lamination methods cannot produce fine patterns or gradients in composition without additional processing. Moreover, aligning the various MEA layers can be difficult.
U.S. Pat. No. 6,187,467 by Zhang et al. discloses the manufacture of electrodes using sequential impregnation and printing. First, a proton conducting polymer is impregnated onto a surface of a substrate, such as carbon fiber paper. The impregnation may be accomplished using a variety of techniques such as dipping the substrate into a solution comprising the proton conducting polymer. An electrocatalyst is then applied to the impregnated substrate. It is disclosed that the electrocatalyst can be applied to the substrate in the form of an aqueous ink including a proton conducting ionomer in solution by any known method including spraying, screen printing and ink-jet printing.
International (WIPO) Publication No. WO 0205365 by Gascoyne et al. discloses an ink composition useful in fabricating the anode of a PEMFC. The ink comprises a liquid medium, which may be aqueous or organic, one or more electrocatalyst, one or more proton-conducting polymer and one or more water retaining materials. It is disclosed that the ink may be applied to a substrate by any variety of methods known in the art such as filtration, vacuum deposition, spray deposition, casting, extrusion, rolling, printing or decal transfer. The substrate may be either the fluid distribution layer or the PEM.
Methods to produce MEAs using ink compositions containing carbon particles and zero-valent Pt compounds have been disclosed by Starz et al. in U.S. Pat. No. 6,500,217. This patent discloses a process for applying electrode layers to the front and back of a polymer electrolyte membrane strip.
Ink-jet deposition of platinum sols has been disclosed by Shah et al. (Langmuir, 1999, Vol. 15, pp. 1584-1587) for use as a catalyst in electroless copper deposition for microelectronic applications.
U.S. Pat. No. 5,672,439 by Wilkinson et al. discloses an MEA for a DMFC wherein crossover of the liquid methanol from the anode to the cathode is reduced. To reduce crossover, catalyst particles that promote the oxidation of the methanol in the anode can be concentrated on a surface of the anode.
U.S. Pat. No. 6,024,848 by Dufner et al. discloses an electrochemical cell such as a PEMFC. The cell can include a bi-layer for enhancing the transport of fluids within the cell. The bi-layer is in contact with an electrode and includes a hydrophobic phase of carbon black and a hydrophobic polymer, as well as a hydrophilic phase including a mixture of carbon black and a proton exchange resin. The bi-layer can be formed by a filter transfer process.
U.S. Pat. No. 6,156,449 by Zuber et al. discloses a catalyst layer, such as in a PEMFC, which includes a proton-conducting polymer, electrically conductive carbon particles and fine particles of a precious metal such as platinum. The layer is formed from an ink composition that includes carbon particles and at least one organic precious metal complex compound in a solution of the ionomer. The composition is applied and dried wherein the complex compounds are thermally decomposed during drying to form precious metal particles.
There is a need for a manufacturing process which can rapidly and economically produce MEAs with the aforementioned transport characteristics. That is where: methanol is transported within the MEA directly to the active sites within the electrocatalyst layer but away from the PEM to prevent cross-over; electrons are transported quickly from the anode electrocatalyst layer to the electrically conductive portions of the anode bipolar plate; water is transported through the anode and PEM quickly to deliver methanol to the active sites, maintain the hydration of the PEM and prevent back-pressure on the PEM; and CO2 is transported out of the anode in its dissolved state to prevent the removal of the methanol fuel.
It would be advantageous to provide an MEA manufacturing process which is capable of depositing all desired materials on fragile or durable substrates in a desired pattern and in a desired concentration. It would be advantageous to provide a single deposition method which is capable of depositing all materials in desired patterns and concentrations. It would be advantageous to provide an MEA manufacturing process which is capable of depositing materials quickly and without substantial changes in materials or downtime. The MEA manufacturing process for a DMFC should be able to construct liquid diffusion electrodes for the DMFC anode and gas diffusion electrodes for the cathode. The design elements of the DMFC gas diffusion cathode can also be incorporated into the design of a gas diffusion anode and therefore used for other types of fuel cells including PEMFC.